Mul i-p oduc s a egy o enhance he en i onmen al
p o ile o he canning indus y owa ds ci cula economy
An onio Co és, Xa ie Es e e-Llo ens, Sa a González-Ga cía, Ma ia Te esa Mo ei a, Gume sindo
Feijoo
Accep ed manusc ip
How o ci e:
Science o he To al En i onmen , 791 (2021), 148249.
h ps://doi.o g/10.1016/j.sci o en .2021.148249
Copy igh in o ma ion:
© 2021 Else ie B.V. This manusc ip e sion is made a ailable unde he CC-BY-NC-ND 4.0
license (h p://c ea i ecommons.o g/licenses/by-nc-nd/4.0/)
1
Mul i-p oduc s a egy o enhance he en i onmen al p o ile o he canning indus y
1
owa ds ci cula economy
2
An onio Co és*, Xa ie Es e e-Llo ens, Sa a González-Ga cía, Ma ia Te esa Mo ei a
3
and Gume sindo Feijoo
4
CRETUS. Depa men o Chemical Enginee ing, Uni e sidade de San iago de Compos ela,
5
15705 San iago de Compos ela (Spain)
6
* Co esponding au ho
7
E-mail. an oniojose.co es.m[email p o ec ed]; an oniojose.co es.mon [email protected]
8
Abs ac
9
The sus ainable and con inued p oduc ion o enough ood o eed he en i e wo ld´s popula ion is
10
one o he main conce ns in he ood indus y. Spain, and in pa icula Galicia, which is an
11
eminen ly ishing egion cha ac e ised by he consump ion o la ge quan i ies o ish, bo h esh
12
and p ocessed, mus ace he challenge o shi ing i s sea ood p oduc i e ab ic owa ds a ci cula
13
economy. To achie e his objec i e, he i s ask is o demons a e ha ci cula economy
14
p inciples allow o educe he en i onmen al impac s associa ed wi h sea ood p oduc ion. In his
15
sense, his s udy p oposes he en i onmen al e alua ion o he skipjack una (Ka suwonus
16
pelamis) alue chain wi hin a canning indus y loca ed in Galicia h ough he LCA me hodology
17
om an a ibu ional pe spec i e, including he alo isa ion p ocesses o biowas e (edible and
18
inedible by-p oduc s). Resul s indica e ha he main c ucial subsys ems o he alue chain a e
19
una ishing and he canning p ocess, as i was expec ed conside ing o he simila s udies on
20
sea ood p oduc s. Mo eo e , his speci ic case s udy demons a es ha he mul i-p oduc s a egy
21
applied o he canning sec o is en i onmen ally iable. Thus, al hough he en i onmen al impac s
22
o he en i e sys em a e inc eased by including u he alo isa ion ope a ions, he en i onmen al
23
loads assigned o he main p oduc (canned una) dec ease compa ed o he one-p oduc sys em
24
by assigning en i onmen al bu dens o o he alue-added p oduc s ( una pâ é, ishmeal, and ish
25
oil).
26
2
Keywo ds
27
Li e Cycle Assessmen ; Canned una; Value chain; Valo isa ion; By-p oduc s
28
29
3
1. In oduc ion
30
As he wo ld´s popula ion has been expanded, he demand o ood and ene gy has seen a apid
31
inc ease. In ac , all p ojec ions indica e ha a leas a 70% inc ease in ood p oduc ion will be
32
needed o mee ood demand by 2050 (FAO, 2012), which is expec ed based on inc eased yields
33
and p oduc i i y o c ops, li es ock and ishe ies. Food p oduc ion is ecognized as a majo
34
con ibu o o en i onmen al impac s in bo h de eloped and de eloping coun ies (Nemecek e
35
al., 2016), amoun ing o a ound 13.7 billion me ic ons o CO2 eq (Poo e and Nemecek, 2018),
36
which ep esen 26% o global an h opogenic g eenhouse gas (GHG) emissions (Pa ke e al.,
37
2018). Del ing deepe in o he key d i e s o his high en i onmen al impac , in addi ion o he
38
in insic impac s o ood p oduc ion i sel , o he "a oidable" impac s play a majo ole, such as
39
he en i onmen al bu dens ela ed o ood packaging and dis ibu ion wo ldwide (Yokokawa e
40
al., 2018). In a global ma ke , he consump ion o some p oduc s p esen s a la ge impac when
41
conside ing he en i e p oduc ion chain om a li e-cycle pe spec i e (c adle- o-pla e app oach).
42
In pa icula , ocusing on he ishe ies sec o , he si ua ion in he oceans is agonizing, wi h ish
43
s ocks being decima ed o e he yea s all o e he wo ld (Wilson e al., 2020). The s a e o he
44
oceans is becoming ex emely wo ying o e ime. In 2017, he maximum peak o o e ished
45
ma ine s ocks (34.2%) and a minimum o unde ished s ocks (6.2%) was eached, acco ding o
46
he esul s published in FAO (2020). In pa allel o he inc easing ise o o e ishing, aquacul u e
47
con inues o g ow s eadily, o he poin ha oday ish p oduced in aquacul u e acili ies accoun
48
o 46% o o al ish p oduc ion (FAO, 2020). A his poin , an in ense deba e has s a ed o eme ge
49
ega ding he long- e m sus ainabili y o wild ishe ies o whe he aquacul u e should be chosen
50
as he main ish sou ce (Ruiz-Salmón e al., 2021). The alo isa ion o was e and disca d ac ions
51
o he p oduc ion o ishmeal and ish oil o be used o he o mula ion o eed o a med ish
52
also needs o be conside ed (F éon e al., 2014c). The e is g owing e idence ha he app oach o
53
u ilize such ac ions: ish bones, isce a, heads and o he less desi able pa s as aw ma e ial o
54
he p oduc ion o alue-added p oduc s such as omega-3 acids and collagen, al hough hese
55
al e na i es a e a a less de eloped s age o indus ial implemen a ion (Laso e al., 2018a).
56
4
In Spain, a coun y ha has adi ionally been an impo an ishing na ion om he poin o iew
57
o ca ching, p ocessing and consump ion (Vázquez-Rowe e al., 2014), 922,564 ons o ish and
58
sea ood we e landed in 2018, making Spain he i s coun y in he Eu opean Union, bo h in e ms
59
o olume and alue, wi h almos 2,150 million eu os (Eu opean Commission, 2020a). Spain has
60
also de eloped an impo an sea ood p ocessing sec o , especially smoked, p ocessed and, abo e
61
all, canned sea ood. Domes ic canned una p oduc ion leads EU p oduc ion, accoun ing o
62
app oxima ely 70% o he o al olume (Ga cía-del-Hoyo e al., 2017). Speci ically, he olume
63
o canned sea ood p oduc ion by Spanish companies eached 353,000 ons in 2018, being Galicia
64
(NW Spain) he leading egion a na ional le el, accoun ing o mo e han 85% o Spanish
65
p oduc ion (EUMOFA, 2019).
66
Galicia´s canning adi ion means ha i is home o 7 o he 10 la ges companies on he Spanish
67
canning sec o , including he Top-5 (A dán, 2018). Mo eo e , he p esence o small and medium-
68
sized en e p ises (SMEs) is p edominan in he Galician canning sec o , wi h a high pe cen age
69
o small companies (<50 employees), which ep esen 66% o he o al, e en highligh ing ha
70
22% o he o al numbe o companies a e e y small wi h less han 10 employees (A dán, 2018).
71
Nume ous ini ia i es and p ojec s o he de elopmen o ci cula economy s a egies a e al eady
72
unde way in di e en companies. Howe e , due o he a o emen ioned cha ac e is ics o he
73
Galician canning indus y, i is di icul o hese ini ia i es o pe mea e he ma ke and in he
74
p esen con ex he Galician canning sec o mus ace he challenge o he cu en pa adigm shi
75
om a linea economy o a ci cula economy in which he main objec i e o companies mus be
76
o maximize p oduc ion h ough he alo isa ion o was e (Ciccullo e al., 2021). In his sense,
77
among he ac ions o be de eloped o achie e a comple e in eg a ion o he ci cula economy
78
wi hin he canning sec o , he ollowing s and ou : (i) he use o p ocessing echniques wi h a low
79
en i onmen al impac ; (ii) he educ ion o he packaging esidues; (iii) he alo isa ion o
80
was ewa e lows; and (i ) he accomplishmen o he objec i e o ze o biological was e,
81
alo ising all biowas e ac ions o p oduce new alue-added p oduc s. Wi h his in mind, he
82
canning indus y in gene al, and he Galician canning indus y in pa icula , has eno mous oom
83
5
o imp o emen , since a la ge pa o he ish is di ec ly disca ded (heads, isce a, bones, e c.),
84
which can open he doo o he de elopmen o new p oduc s (Ga cía-San iago e al., 2020).
85
Ci cula economy eme ges as an opposi e solu ion o he cu en linea sys em as a sus ainable
86
sys em whe e economic g ow h is decoupled om esou ces use, h ough he educ ion in he
87
consump ion and he eci cula ion o aw ma e ials (Ko honen e al., 2018). On he main poin s
88
o he ci cula economy is he educ ion o was e gene a ed h oughou he alue chain, alo ising
89
hem as aw ma e ials o he gene a ion o added- alue p oduc s. Howe e , inc easing ci cula i y
90
does no necessa ily ansla e in o a di ec educ ion o en i onmen al impac s (Nie o and Kalba ,
91
2019), which c ea es a dilemma o decision-make s when selec ing adequa e ci cula p ac ices
92
and inno a ions (Ru í-Salís e al., 2021). The impac s o bene i s gene a ed by hese ci cula
93
s a egies a e o en measu ed h ough he use o ci cula i y me ics (Co ona e al., 2019). Li e
94
Cycle Assessmen (LCA) me hodology, as i is based on he quan i ica ion o he inpu s and
95
ou pu s o a sys em, becomes a good example o a ci cula i y assessmen ool o quan i y and
96
e alua e he bene i s o impac s o ci cula economy s a egies. The applica ion o LCA
97
me hodology o de e mine he en i onmen al impac s o ish ca ches, a ming (aquacul u e), and
98
p ocessing s a ed in he mid-2000s. A long lis o LCA sea ood s udies on di e se pelagic species
99
such as ho se macke el (Vázquez-Rowe e al., 2010), Pe u ian and Can ab ian ancho y (F éon e
100
al., 2014b; Laso e al., 2018b), ca p (Ho nbo g and F ämbe g, 2020) o A lan ic macke el (Ramos
101
e al., 2011) ha e been epo ed. Deme sal species such as hake (A adí e al., 2018; Vázquez-
102
Rowe e al., 2011b), cod (S anes e al., 2011; Ziegle e al., 2013) o oc opus (Vázquez-Rowe e
103
al., 2012b), c us acean species such as p awns (Fa me y e al., 2015; Medei os e al., 2017),
104
lobs e (D iscoll e al., 2015) o goose ba nacle (Vázquez-Rowe e al., 2013a), and bi al e species
105
such as A lan ic scallop (Co és e al., 2021) ha e been epo ed. Rega ding aquacul u e, di e en
106
s udies on mussels (I iba en e al., 2010b, 2010c; Lou guioui e al., 2017; Tambu ini e al., 2020),
107
oys e s (Tambu ini e al., 2019), u bo (I iba en e al., 2012) o salmon a ming (Philis e al.,
108
2021) can be highligh ed. I is also impo an o men ion ha adi ionally no only ishing o
109
a ming ac i i ies ha e been e alua ed, bu also he p oduc ion o di e en ish- and sea ood-
110
6
based p oduc s such as ish s icks (Vázquez-Rowe e al., 2013b), ishmeal and ish oil (F éon e
111
al., 2017) and canned p oduc s (Almeida e al., 2015; A adí e al., 2015, 2014; I iba en e al.,
112
2010a; Laso e al., 2017; Vázquez-Rowe e al., 2014). Re iew a icles on ishing (A adí and
113
F éon, 2013), aquacul u e (Bohnes e al., 2019; Bohnes and Lau en , 2019; Philis e al., 2019) and
114
p ocessing (Ruiz-Salmón e al., 2021; Vázquez-Rowe e al., 2012a) s ages ha e also been
115
e alua ed.
116
This s udy p oposes he en i onmen al e alua ion o he skipjack una (Ka suwonus pelamis)
117
alue chain wi hin a canning indus y loca ed in Galicia h ough he LCA me hodology. The
118
p oduc ion line ocuses on gou me p oduc s, wi h high added alue, basing i s p oduc ion on
119
local aw ma e ials, wi h adi ional manu ac u ing me hods and using, as a as possible, ce i ied
120
o ganic ing edien s. Thus, he p ocessing plan mee s some o he ci cula economy p inciples: (i)
121
he ish is caugh wi h adi ional echniques in na ional ishing g ounds; (ii) adi ional echniques
122
such as cooking in seawa e and ai -d ying a e ollowed; (iii) he p ima y packaging is made o
123
aluminium, so i is 100% ecyclable; (i ) high quali y by-p oduc s (non-canned edible pa s) a e
124
used o p oduce o he p oduc s; ( ) he low-quali y by-p oduc s a e alo ised in he o m o
125
ishmeal ha could be used o animal eed. In his way, he canning plan minimises he
126
consump ion o aw ma e ials, minimises anspo and ollows a mul i-p oduc s a egy. The
127
main objec i es o he s udy a e o de e mine he en i onmen al iabili y o his app oach and o
128
lay he ounda ions o he way o wa d o o he companies o posi ion hemsel es in a highly
129
compe i i e ma ke . The main no el y o his s udy lies in he ac ha he LCA me hodology has
130
been used o analyse he en i onmen al impac s o he en i e canned una alue chain, and no
131
only hose impac s assigned o he p oduc ion o he main p oduc . The inclusion o he
132
alo isa ion p ocesses o esidual o ganic ac ions wi hin he sys em bounda ies makes i possible
133
o analyse he p oduc om a b oade poin o iew and opens he doo o he iden i ica ion and
134
e alua ion o oppo uni ies o en i onmen al imp o emen .
135
2. Ma e ials and me hods
136
2.1. De ining he goal and scope. Impac assessmen me hodology.
137
7
Mo ing owa ds a e i able ci cula economy equi es aking small s eps o demons a e he
138
iabili y o mul i-p oduc p ocesses om an en i onmen al poin o iew. In his sense, his s udy
139
aims o assess he en i onmen al sus ainabili y o he en i e canned una alue chain ollowing
140
he Li e Cycle Assessmen me hodology (ISO 14040; 14044) om a a ibu ional pe spec i e.
141
Al hough he main p oduc is canned una, all s ages o he alue chain we e included wi hin he
142
sys em bounda ies, including he manu ac u e o by-p oduc s and he alo isa ion o o ganic
143
was e. Thus, he main objec i e o his s udy is o de e mine om an en i onmen al poin o iew
144
whe he he p oduc ion o mul iple alue-added p oduc s is mo e sus ainable han single-p oduc
145
app oaches.
146
A c adle- o-ga e app oach was conside ed in he s udy, ha is, conside ing he ex ac ion o aw
147
ma e ials o p oduce he equi ed inpu s and he manu ac u e o he p oduc s, bu no he
148
consump ion and inal disposal s ages. This pe spec i e was assumed since he main objec i e o
149
he s udy is o ecognize he en i onmen al implica ions o he p oduc ion o una-based p oduc s.
150
The main aw ma e ial is skipjack una, so ishing and anspo o he canning plan , as well as
151
he p oduc ion and anspo o o he ing edien s and packaging ma e ials, we e included in he
152
sys em bounda ies. The Func ional Uni (FU) conside ed o assessmen was 1 onne o aw una
153
a p ocessing plan ga e since i seems consis en o selec a eeds ock-based FU as he plan is
154
cha ac e ized by i s mul i-p oduc na u e. The so wa e SimaP o 9.0 (PRe-Consul an s, 2017) was
155
used o he compu a ional implemen a ion o he in en o ies. The li e cycle impac assessmen
156
s ep was ca ied ou using he ReCiPe 2016 1.1 me hodology in a hie a chis pe spec i e a
157
midpoin le el (Huijb eg s e al., 2017). The en i onmen al bu dens we e calcula ed in e ms o
158
he ollowing impac ca ego ies: Global Wa ming (GW), S a osphe ic Ozone Deple ion (SOD),
159
Te es ial Acidi ica ion (TA), F eshwa e Eu ophica ion (FE), Ma ine Eu ophica ion (ME),
160
F eshwa e Eco oxici y (FET), Ma ine Eco oxici y (MET), Mine al Resou ces Sca ci y (MRS)
161
and Fossil Resou ces Sca ci y (FRS).
162
2.2. Desc ip ion o he sys em unde s udy.
163
8
The alue chain associa ed wi h canned una was di ided in 4 di e en subsys ems, as depic ed
164
in Figu e 1. Subsys em 1 is ela ed o he ishing and anspo a ion o una as he main aw
165
ma e ial o supply he canning plan . Subsys ems 2-4 a e linked o he di e en ac i i ies and
166
ope a ions ha ake place wi hin he canning ac o y. I is impo an o no e ha he aim is o use
167
he esidual ac ions o he p ocess o p oduce alue-added p oduc s; howe e , he ea men o
168
non- eco e able was e and was ewa e has been included in all subsys ems. All liquid ac ions
169
a e di ec ly sen o a municipal was ewa e ea men plan loca ed close o he si e. On he o he
170
hand, packaging was e is ecycled as a as possible and land illed o incine a ed acco ding o he
171
Spanish p o ile.
172
173
Figu e 1. Sys em bounda ies o he en i onmen al assessmen o canned una alue chain.
174
15
Rowe e al., 2011b), and bai ishing and p ocessing (Vázquez-Rowe e al., 2014). As can be
305
obse ed in Figu e 3.a, uel p oduc ion and consump ion du ing ishing ope a ions u ned ou o
306
be one o he main sou ces o en i onmen al impac , accoun ing o mo e han 83% o he o al
307
impac in he TA ca ego y, 78% in FRS and 65% in GW. This esul was expec ed and is in line
308
wi h o he ishing lee s e ised, whe e di ec and indi ec uel emissions we e highligh ed as he
309
mos impo an ca ie o GHG emissions (A adí e al., 2018; Sandison e al., 2021; Vázquez-
310
Rowe e al., 2014; Villanue a-Rey e al., 2018), as well as o he en i onmen al impac s, such as
311
e es ial acidi ica ion (Ziegle e al., 2016). In his a icle, a Fuel Use In ensi y (FUI) o 548.9
312
L/ onne was ob ained o pole-and-line una ishing. This esul can be compa ed wi h di e en
313
esul s published in scien i ic li e a u e; e.g. Mille e al. (2017) es ima ed he uel consump ion
314
o he Maldi ian pole-and-line lee among di e en shoal o una, a ying om 200 L/ onne o
315
almos 600 L/ onne.
316
I is wo h no ing ha special emphasis is placed on he ac ha bai accoun s o app oxima ely
317
15-20% o he amoun o uel consumed in a ishing ip, which coincides wi h he esul s ob ained
318
in his s udy, since he impac s associa ed wi h ishing and bai p oduc ion each 71.6% in ME
319
and do no all below 16% in any impac ca ego y. Pole-and-line una ishing me hods ha e no
320
been adi ionally s udied, so he e a e only a ew s udies ha quan i y he uel consump ion o
321
his ype o ishing gea wi h di e en a ge species, howe e , una ishing by di e en ishing
322
gea s has been ex ensi ely s udied. In his con ex , Hospido and Tyedme s (2005) quan i ied he
323
uel consump ion o he Spanish una pu se seine ishing lee in he Indian (373 L/ onne), A lan ic
324
(442 L/ onne) and Paci ic (442 L/ onne) oceans, se ing he amewo k o he quan i ica ion o
325
FUI o o he ishe ies in he u u e. Mo eo e , Pa ke e al., (2015) achie ed a signi ican ly la ge
326
sample o essels o upda e his esul , ob aining on a e age ha he una pu se seine ishe y
327
consumes on a e age 365 L/ onne.
328
16
Speci ically, he pu se seine ishe y o skipjack una in A lan ic wa e s equi es a uel
329
consump ion o 445 L/ onne acco ding o hei esul s. On he o he hand, Pa ke and Tyedme s
330
(2015) es ima ed a uel consump ion o 1,612 L/ onne o la ge pelagic (mainly una) ishe ies
331
using longlines and o he o ms o pole-and-lines. F om ano he , much mo e gene ic poin o
332
iew, Pa ke e al. (2018) calcula ed he global CO2 emissions linked o uel combus ion in ishing
333
essels, es ima ing he FUI o pelagic ish (> 30 cm) a 430 L/ onne. In b ie , mos o he alues
334
p o ided o una ishing, excep o longlines, a e wi hin he same ange, which can gi e he idea
335
ha he alues ob ained in his s udy a e close o eali y. Howe e , i is impo an o no e ha FUI
336
measu emen s om p e ious s udies may a y conside ably depending on bo h he measu emen
337
me hod and he analysed ishing gea (Pa ke e al., 2015).
338
On he o he side, he p oduc ion and consump ion o an i ouling has been e ealed as a
339
di e en ial elemen in he eco oxici y ca ego ies (74.5% in FET and 71% in MET) due o coppe
340
and zinc emissions du ing he use s age, as demons a ed in p e ious li e a u e (A adí and F éon,
341
2013). I is also no ewo hy he 20% o o al ishing impac in MRS ca ego y, due o he ex ac ion
342
o bauxi e o p oduce coope needed o an i ouling o mula ion. The bauxi e o e is ea ed wi h
343
dilu e sulphu ic acid o e a pe iod o mon hs, dissol ing coppe o o m a weak solu ion o coppe
344
sulpha e, om which coppe can eco e ed by elec olysis. Ice p oduc ion con ibu es much less
345
o he en i onmen al bu dens o he sys em, i s con ibu ion emains almos cons an in all
346
ca ego ies anging om a minimum o 0.5% o a maximum o 2%, excep in he FET ca ego y,
347
whe e i eaches a maximum o 7%, mainly due o elec ici y consump ion and he high
348
dependence on coal in he Spanish elec ici y p o ile. Wi hin he GW ca ego y, he p esence o
349
R-410A s ands ou , which is a e ige an gas widely used in e ige a ion machines and whose
350
leakage in luences he ca bon oo p in o he p ocess, as has been shown in p e ious li e a u e
351
(Vázquez-Rowe e al., 2011b). Finally, he en i onmen al impac o pain consump ion is
352
p ac ically negligible, while he ea men o he was e (bo h solid and liquid) is only ema kable
353
in he ME ca ego y due o he ea men o bilge was ewa e .
354
17
355
Figu e 3. Rela i e con ibu ion o en i onmen al impac s associa ed wi h Subsys em 1. Tuna
356
ishe y (a) and Subsys em 2. Canning p ocess (b).
357
Acco ding o he esul s shown in Figu e 3.b, he la ges con ibu ion o he en i onmen al impac
358
in 6 o he 9 impac ca ego ies analysed comes om he p oduc ion o aluminium p ima y
359
packaging, p o ing o be one o he mos impo an elemen s in he p oduc ion o canned sea ood
360
18
(Almeida e al., 2014; Hospido e al., 2006). In ac , he p oduc ion o aluminium cans is he main
361
con ibu o o he MRS ca ego y, whe e i eaches he highes con ibu ion o any impac ca ego y
362
(95%), linked o he p oduc ion o i gin aluminium, which equi es a high consump ion o
363
bauxi e (almos 5 kg pe kg o aluminium) (Laso e al., 2017). Hea p oduc ion by na u al gas ( o
364
cooking and s e ilisa ion) is also an impo an con ibu o o GW (36.2%) and FRS (43%),
365
al hough i s con ibu ion o he o he ca ego ies is in all cases less han 5%. Was ewa e ea men ,
366
as in he ishe ies subsys em (Figu e 3.a) is only ele an in he ME ca ego y (50%) due o
367
nu ien emissions o he ea ed e luen , while in he o he ca ego ies, i ha dly exceeds 6%. The
368
p oduc ion o seconda y packaging (ca dboa d and plas ic) p esen s a ela i e cons an
369
con ibu ion a ound 5-7%, eaching 12.2% in FE and d opping o 2.9% in MRS. Elec ici y
370
consump ion is almos negligible, only ele an in he FE and TA ca ego ies (9.2% and 8.3%,
371
espec i ely), while in he es o he ca ego ies i ha dly exceeds 4%. Finally, he p oduc ion o
372
he ag icul u al ing edien s o he ga nish and seasoning (onion, ga lic, peppe , e c.) and he
373
anspo o was e o specialised plan s ha e a e y low con ibu ion in all impac ca ego ies,
374
always below 0.5%. The nume ical esul s ela ed o he FU can be ound in he Table S.4 o he
375
supplemen a y ma e ial.
376
3.2. E ec o he alloca ion s a egies on he en i onmen al p o ile
377
This s udy has assessed he en i e canned una alue chain om an en i onmen al poin o iew,
378
om he p oduc ion o aw ma e ials o he p ocessing o p oduc s and co-p oduc s and he
379
ea men o was e, ying o mo e owa ds he a ge o 0 bio-was e. This s udy is an example o
380
sys em expansion o a oid he use o alloca ion s a egies, as p io i ised in he ISO s anda ds.
381
Howe e , when he objec i e o he s udy is o analyse he en i onmen al impac s associa ed wi h
382
a pa icula p oduc wi hin he alue chain, and ull seg ega ion o ma e ial and ene gy
383
consump ion o each p oduc ion line is no possible, he use o alloca ion ac o s o accu a ely
384
epo en i onmen al bu dens seems una oidable (Aye e al., 2007). T adi ionally, he selec ion
385
o alloca ion ac o s is one o he p ocedu es ha gene a es he leas consensus among LCA
386
p ac i ione s. In his ega d, di e en au ho s ha e p oposed di e en alloca ion me hods o
387
19
di e en case s udies. Wi hin he sea ood-speci ic case s udies, o example, Th ane (2006) used
388
sys em expansion s a egies o handle he co-p oduc alloca ion in he LCA o di e en ish
389
p oduc s. On he o he side, Ziegle e al. (2003) applied economic alloca ion ac o s o calcula e
390
he en i onmen al bu dens o cod ille s, bu also hey applied mass alloca ion conside ing ha
391
p ice luc ua ions may condi ion he eliabili y o he esul s. On his basis and conside ing ha
392
he main objec i e o he canning indus y is he p oduc ion o ma ke able p oduc s ha gene a e
393
an economic income, he economic alloca ion was calcula ed o canned una (95.8%), una pâ é
394
(3.3%), ishmeal (0.7%) and ish oil (0.2%).
395
When analysing he en i onmen al p o ile o he main p oduc (canned una), one di e en
396
app oach om sys em expansion co esponds o he exclusion o subsys ems SS3 and SS4 om
397
he sys em bounda ies, including wi hin he bounda ies o SS2 he o ganic was e ea men
398
p ocesses co esponding o he co-p oduc s (635 kg pe 1,000 kg o aw una). Howe e , his
399
app oach would no be en i ely ealis ic, as cu en ly, he p ocessing o ish co-p oduc s in o
400
ishmeal and ish oil o eed o mula ion is a widely used op ion o he ea men o bio-was e.
401
In any case, Figu e 4 shows he a ia ion in he en i onmen al p o ile o he o iginal alue chain
402
(Base scena io) compa ed o he scena io in which only una ishing and p ocessing o p oduced
403
canned una is included, bu he alo isa ion o o ganic was e om he p ocessing is excluded
404
(Non-ci cula scena io).
405
406
0,8
0,85
0,9
0,95
1GW
SOD
TA
FE
MEFET
MET
MRS
FRS
Base scena io Non-ci cula scena io
20
Figu e 4. Compa a i e en i onmen al p o ile o he wo al e na i e canned una p oduc ion
407
scena ios.
408
I seems ob ious ha emo ing SS3 and SS4 wi h all associa ed ma e ial and ene gy consump ion
409
om he sys em bounda ies will educe he en i onmen al impac o he en i e alue chain.
410
Especially ele an is he educ ion in he ca ego ies ME and FE, whe e he con ibu ion o he
411
SS4 subsys em was highe due o he en i onmen al bu dens o was ewa e ea men , as can be
412
seen in Figu e 2. Howe e , i is ema kable ha he impac o he non-ci cula app oach is almos
413
equal o ha o he baseline scena io in he ca ego ies o MET, FET and TA. E en mo e
414
ema kable is ha he impac on he SOD ca ego y is highe in he non-ci cula scena io, as he
415
emissions om bio-was e ea men a e highe han he emissions om he inpu s and ou pu s o
416
SS3 and SS4.
417
In o de o ela i ise he impac s o he sys em owa ds he p oduc ion o canned una, in he case
418
o he baseline scena io i is necessa y o apply economic alloca ion ac o s, while o he non-
419
ci cula scena io, he en i e en i onmen al bu den is alloca ed o canned una, as i is he main
420
ou pu o he sys em. Conside ing ha only 365 kg ou o 1,000 kg a e canned and ha 90 g o
421
una and 30 g o addi i e a e pu in o each can, he ca bon oo p in o a can o una was quan i ied
422
as 0.98 kg CO2 eq in he baseline scena io (conside ing he economic alloca ion o all co-
423
p oduc s), while in he non-ci cula scena io i eached 1.01 kg CO2 eq pe can. To gi e a b oade
424
alue and no so ocused on he ca bon oo p in , he o al en i onmen al impac s we e calcula ed
425
in e ms o he ReCiPe me hodology endpoin , anging be ween 0.050 and 0.052 p s pe can in
426
he baseline scena io and in he non-ci cula scena io, espec i ely.
427
In his way, he compa ison be ween he en i onmen al p o ile o he sys em e alua ed in Figu e
428
1 (Base scena io) and his new scena io (Non-ci cula scena io) allow o answe he ques ion: is
429
he applica ion o mul i-p oduc s a egies en i onmen ally iable when he objec i e o he
430
assessmen is o assess only one o he p oduc s, wi hou conside ing he p oduc s a oided and
431
he en i onmen al c edi s? In he speci ic case s udied and aking in o accoun ha he inclusion
432
o SS3 and SS4 has no had much in luence on he o al impac o he sys em (less han 6% on
433
21
a e age), he dis ibu ion o he loads be ween he 4 di e en p oduc s has made i possible o
434
educe he li e cycle impac o canned una, while he en i onmen al pe o mance o he whole
435
alue chain is ha dly modi ied. I can be seen ha a ci cula economy app oach is easible and
436
e ec i e and hese p inciples a e in line wi h hose men ioned by he Eu opean Commission in
437
he Ci cula Economy ac ion plan (Eu opean Commission, 2020b).
438
3.3. Benchma king wi h o he canned sea ood
439
When compa ing wi h o he esul s a ailable in he pee - e iewed li e a u e, i is impo an o
440
no e ha he e a e some issues ha need o be add essed in de ail in o de o make ealis ic
441
compa isons (A adí and F éon, 2013; Vázquez-Rowe e al., 2012a): (i) The li e cycle impac
442
me hodology should be de ailed since, al hough he ca ego ies o di e en me hodologies may be
443
analogous, in many cases he uni s o measu emen a e di e en ; (ii) The unc ional uni selec ed
444
o he analysis, as well as he edible con en o each p oduc , o adequa ely es ima e he associa ed
445
en i onmen al impac pe uni and pe quan i y o p oduc (e.g. 1 kg); and (iii) he en i onmen al
446
indica o used o compa ison (en i onmen al oo p in , no malised impac ac o , e c.). Taking
447
all his in o ma ion in o accoun , Table 3 shows he de ailed esul s o he compa ison be ween
448
he ca bon oo p in o di e en canned sea ood p oduc s.
449
450
22
Table 3. Ca bon oo p in alues o o he canned sea ood p oduc s
451
P oduc
Assessmen me hod
P oduc
kg CO2/kg
Re e ence
Tuna
ReCiPe 2016
90 g una
30 g addi i e
20 g can
5 g boa d
8.2
P esen s udy
Pilcha d
CML-IA Baseline
95 g pilcha d
35 g oli e oil
20 g can
7.5
(Almeida e al.,
2015)
Pilcha d
ReCiPe 2008
85 g pilcha d
35 g oli e oil
20 g can
25.2
(Vázquez-Rowe e
al., 2014)
Mussels
IPCC 2016
129 g mussels
120 g sauce
81 g can
12.7 g boa d
17.5
(I iba en e al.,
2010a)
Tuna
ReCiPe 2008
n.d.
3.7
(A adí e al., 2015)
Pe u ian
ancho y
ReCiPe 2008
n.d.
1.7
(A adí e al., 2014)
Can ab ian
ancho y
ESA wi h me ics om
ICheme 2002
30 g ancho y
20 g oli e oil
15 g can
5 g boa d
4.7
(Laso e al., 2017)
452
Is impo an o no e ha Table 3 compiles he esul s wi h a c adle- o-ga e app oach, conside ing
453
he impac s ela ed o ishing, p ocessing and packaging. Thus, in hose s udies ha conside ed
454
he dis ibu ion and consump ion s ages (Almeida e al., 2015; Laso e al., 2017; Vázquez-Rowe
455
e al., 2014), he en i onmen al bu dens co esponding o hese s ages we e no aken in o accoun .
456
The ca bon oo p in o canned una p esen s an in e media e alue, much lowe han he alues
457
p esen ed by sa dine in oli e oil (Vázquez-Rowe e al., 2014) and mussels, mainly due o he
458
p oduc ion o he p ima y packaging, as hese wo cases we e packed in inpla e. Howe e , his
459
alue is e y simila o he associa ed wi h he can o sa dines wi h oli e oil assessed in Almeida
460
e al. (2015), whe e simila packaging is used and simila echniques a e ollowed.
461
Can ab ian ancho y is also packaged in aluminium and, despi e ha ing a much lowe ish/package
462
a io han ha ob ained in his s udy, hey epo ed only 4.7 kg CO2/kg. This is because he
463
p ocessing ope a ions a e e y di e en , and he ca ch a ios o his ishe y a e much highe . In
464
his sense, he low esul s o Ecuado ian canned una (A adí e al., 2015) and Pe u ian ancho y
465
23
(A adí e al., 2014) can be explained by lowe uel use in he Ecuado ian and Pe u ian ishe ies,
466
mainly due o a be e ca ch pe uni e o in ela ion o a highe abundance o he esou ce (F éon
467
e al., 2014a). While i is ue ha in all cases i was concluded ha bo h he ishing s age and he
468
p oduc ion o p ima y packaging ( inpla e o aluminium) a e he main d i e s o en i onmen al
469
impac s and all imp o emen ac ions should ocus on hem.
470
4. Conclusions
471
This s udy demons a es ha , om a p oduc app oach, he inclusion o by-p oduc alo isa ion
472
p ocesses o add ess a mul i-p oduc s a egy imp o es he en i onmen al p o ile o he main
473
p oduc . I is a clea example o a sys em expansion o a oid bu den alloca ion be ween p oduc s
474
when he ocus is on he assessmen o he en i e alue chain. When he ocus is on he assessmen
475
o a single p oduc , he alloca ion o en i onmen al bu dens seems una oidable.
476
I has been shown ha he ishing and p ima y p ocessing s ages a e he mos ele an sub-sys ems
477
wi hin he en i onmen al p o ile o he canned una alue chain. The in en o y o he ishing s age
478
showed, as p e ious s udies on di e en ishing lee s, ha he impac s o he ishing s age come
479
mainly om he p oduc ion and consump ion o diesel and an i ouling. In his case, he
480
impo ance o he bai used o ishing also s ands ou , as i equi es he ishing and p ocessing o
481
sa dine o use as bai . P ima y packaging p esen ed he highes en i onmen al impac in he li e
482
cycle impac s o canned una. Aluminium p oduc ion, lamina ion and ex usion had he highes
483
impac in almos all impac ca ego ies, as expec ed o canned p oduc s. By-p oduc alo isa ion
484
p ocesses, bo h edible and inedible, ha e p o en o ha e a low impac .
485
This sys em has allowed an app oxima ion o he EU a ge owa ds a c adle- o-c adle sys em
486
app oach, achie ing he goal o ze o biowas e. The esul s show he need o imp o e he
487
applica ion o he ci cula economy in he p ima y sec o , con e ing was e in o aw ma e ials o
488
he p oduc ion o new p oduc s, minimising he consump ion o ma e ial and ene gy esou ces.
489
In his sense, he applica ion o mul i-p oduc s a egies has been shown o imp o e he
490
en i onmen al p o ile o canned p oduc s h ough he alloca ion o en i onmen al bu dens among
491
he new p oduc s; al hough u he analysis om a sus ainabili y poin o iew is equi ed. Simila
492
24
s udies need o be u he applied o speci ic p ima y sub-sec o s in he u u e o con inue he pa h
493
owa ds a mo e sus ainable and ci cula ood sys em.
494
Acknowledgemen s
495
This esea ch was suppo ed by he EAPA_576/2018 NEPTUNUS p ojec . S.G-G would like o
496
exp ess he g a i ude o he Spanish Minis y o Economy and Compe i i eness o inancial
497
suppo (G an e e ence RYC-2014-14984). The au ho s belong o CRETUS and he Galician
498
Compe i i e Resea ch G oup GRC ED431C 2017/29 co- ounded by Xun a de Galicia and
499
FEDER (EU).
500
Re e ences
501
Abdou, K., Gascuel, D., Aubin, J., Romdhane, M.S., Ben Rais Las am, F., Le Loc’h, F., 2018.
502
En i onmen al li e cycle assessmen o sea ood p oduc ion: A case s udy o awle
503
ca ches in Tunisia. Sci. To al En i on. 610–611, 298–307.
504
h ps://doi.o g/10.1016/j.sci o en .2017.08.067
505
Abdou, K., Le Loc’h, F., Gascuel, D., Romdhane, M.S., Aubin, J., Ben Rais Las am, F., 2020.
506
Combining ecosys em indica o s and li e cycle assessmen o en i onmen al assessmen
507
o deme sal awling in Tunisia. In . J. Li e Cycle Assess. 25, 105–119.
508
h ps://doi.o g/10.1007/s11367-019-01651-5
509
Almeida, C., Vaz, S., Cab al, H., Ziegle , F., 2014. En i onmen al assessmen o sa dine
510
(Sa dina pilcha dus) pu se seine ishe y in Po ugal wi h LCA me hodology including
511
biological impac ca ego ies. In . J. Li e Cycle Assess. 19, 297–306.
512
h ps://doi.o g/10.1007/s11367-013-0646-5
513
Almeida, C., Vaz, S., Ziegle , F., 2015. En i onmen al Li e Cycle Assessmen o a Canned
514
Sa dine P oduc om Po ugal. J. Ind. Ecol. 19, 607–617.
515
h ps://doi.o g/10.1111/jiec.12219
516
A dán, 2018. In o me económico y de compe i i idad 2018. Capí ulo 11: El sec o de la pesca
517
en Galicia.
518
A adí, A., Ad ien, R., A amayo, V., F éon, P., 2018. En i onmen al assessmen o he Pe u ian
519
indus ial hake ishe y wi h LCA. In . J. Li e Cycle Assess. 23, 1126–1140.
520
h ps://doi.o g/10.1007/s11367-017-1364-1
521
A adí, A., Bolaños, C., Sando al, I., Ycaza, C., 2015. Li e cycle assessmen o Ecuado ian
522
p ocessed una. In . J. Li e Cycle Assess. 20, 1415–1428. h ps://doi.o g/10.1007/s11367-
523
015-0943-2
524
A adí, A., F éon, P., 2013. Li e cycle assessmen o ishe ies: A e iew o ishe ies scien is s
525
and manage s. Fish. Res. 143, 21–38. h ps://doi.o g/10.1016/j. ish es.2013.01.006
526
A adí, A., F éon, P., Quispe, I., 2014. En i onmen al assessmen o Pe u ian ancho e a ood
527
p oduc s: Is less e ined be e ? In . J. Li e Cycle Assess. 19, 1276–1293.
528
h ps://doi.o g/10.1007/s11367-014-0737-y
529
Aye , N.W., Tyedme s, P.H., Pelle ie , N.L., Sonesson, U., Scholz, A., 2007. Co-p oduc
530
alloca ion in li e cycle assessmen s o sea ood p oduc ion sys ems: Re iew o p oblems
531
and s a egies. In . J. Li e Cycle Assess. 12, 480–487.
532
h ps://doi.o g/10.1065/lca2006.11.284
533
Bohnes, F.A., Hauschild, M.Z., Schlund , J., Lau en , A., 2019. Li e cycle assessmen s o
534
aquacul u e sys ems: a c i ical e iew o epo ed indings wi h ecommenda ions o
535
policy and sys em de elopmen . Re . Aquac. 11, 1061–1079.
536
